The control box
I am making a start here as in some respects it is the easiest subject to tackle once the drive motors have been decided upon.
I confess to a slight cheat on the selection of drive motors as there are others that have trod this path before commercially and they seem to have chosen the AS23 size of stepping motor for all three axis. This size of motor seems to match the AC570764525I motor from Arc Euro Trade that provides 180 Ncm of holding torque and so that is the one I opted to use. Time will tell if it proves to be wrong!
As I intended to have four axis operational A box containing the four stepper motor drivers, their power supply and the the "breakout" board needed to be engineered. As the fourth axis, being the rotary table drive, was not a drive in constant use I chose to make the power supply and driver electrics for this separate to the other three in order for it to be switched off when not in use although housed in the same box.
Power supplies are available commercially but are not difficult to build if one possesses some competence in electrics/electronics. It is not advisable to build your own if this is not an area of expertise that you posses.
Helpfully Arc Euro Trade provide a power supply design on their web site and this design was utilised with one being rated to drive three motors and one rated for the single motor.
Power supply
My supplies utilised a single smoothing capacitor of 10000uF rather than the two shown in parallel
A 5V dc supply is required for the "breakout" PC board and rather than use the supply from the computer I built a regulated 5v dc supply derived from the "Electronics in Meccano" web site.
5V & 12V Power supply circuit
As the fan to be fitted in the box to keep everything cool was an old 12v dc computer fan that I already had, it too needed a dc supply so a 12V regulated dc supply was built based on the same circuitry as the 5v but with a 12V regulator chip.
So they being all the main electric/electronic elements the build could commence.
The box could have been obtained commercially. They are not cheap. I fabricated mine from 1/2" brass angle and aluminium plates for the sides, bottom and top. The sides were 18g and the top and bottom 16g. The whole was held together with 5BA c/s brass screws and nuts. The overall size of the box was 300 x 300 x100 mm. It was not cheap either! I think had aluminium angle been used and steel screws it would have saved money.
Also to be fitted to the box are the on/off switches, Emergency Stop, fuses and cable entries. One advantage of the fabricated box was the ease of removing a side plate or base to drill the holes needed for the various components.
Consideration in the layout of components is needed to accommodate the cables, connecting wires and last but not least the cooling arrangements. My layout can be seen in the photograph with all the cable entries through one side, switches and fuses on another and fan and cooling slots on opposite sides, the fan being on the right-hand side corner in the photo.
Click to enlarge
The main power components are bolted to the base of the box, namely the two torroidal transformers and the rectifiers. The smoothing capacitors are fixed to the base by an epoxy glue. The stepper motor drivers (the black boxes with white writing) are bolted to the base as is the transformer supplying the 12V supply to the 12V and 5V DC supply PCB's. These PCB's are mounted on pillars from the base. The "breakout" board is mounted on stand off brackets from the sides, this can be seen in the top corner of the photo.
A number of the components have tag connections that enable the use of crimped spade terminals. Terminal blocks are mounted on top of the transformers to facilitate their wire connections. The two low voltage DC supply boards and the "breakout" board all have screw terminal connections (except for the D connector on the "breakout" board) so the only soldered connections are for the capacitors.
The wiring is straightforward enough, the points of note were to ensure the mains supply cable could not be pulled out, which was achieved by the simple expedient of a knot in the cable, similarly for the cable from the "breakout" board to the PC. The wires from the breakout board to the motor drivers were a screened cable although the screen was not connected to earth. The wires are: step (or pulse), direction and common (or "near ground").
It is important that when ready to test the electrics that the motor drivers are not in the open circuit condition. The motors should be connected as some motor drivers dislike intensely being switched on with an open circuit and may be irreparably damaged. Alternatively the power side can be checked without motor drivers connected and the breakout board can be checked using an LED and resistor to see the outputs operating when driven from a PC. I found checking the power circuits alone that the surge current is quite high on switch on, probably due to the very large capacitors charging on the secondly side of the transformer and a 13A fuse was necessary in the mains plug. A 10A fuse was blown.
The stepper motors I have fitted have four windings i.e. 8 wires and the windings are connected in a manner depending upon the service the motor is required to perform. The choice is straightforward between maximum torque or maximum speed. For maximum torque the pairs of windings are connected in parallel, and for maximum speed they are connected in series. As the speed is perfectly adequate for home use the maximum torque configuration is adopted. The configuration wiring is actually done at a small terminal box epoxy glued to the motor so that only a four core screened cable comes from the motor to the motor drivers. This four core cable is rated at the full motor current, in my case 2.5 Amps per phase and is fully screened as it carries digital stepping signals at a relatively quite high power and neither radiated noise nor spurious interference is wanted on the motor cables. The cable screen is earthed at the control box only.
Four other connections have to be accommodated, these being the two cores that connect each "limit" microswitch on the X and Y tables to the "breakout" board inputs.
The motor cables and microswitch cable are run in plastic flexible trunking from the control box to the Mill, the microswitch cable is a simple twin bell wire with one connection to the so called "near earth" connection available on the "breakout" board and the other via the limit switch to the input terminal. The limit switch is wired normally closed, i.e. when operated it goes open circuit this being a safe mode of operation.
The emergency stop switch is worth a paragraph in that it is wired as the first item from the mains supply before the ON/OFF switches for the XYZ and A drives. The switch is a a double pole normally closed configuration that goes open circuit when pushed. One pole carries the live supply and the secondary pole carries the contact that the breakout board senses to give a ES signal to the software.
The "breakout" board from CNC4PC is a parallel interface card that connects via a 25D connector cable to the parallel port of the computer. The cable has a male 25D connector at each end. It also needs to be a cable with all pins wired one to one. Some cables with 25D connectors at each end are for data transfer between computers and not all pins are wired. These cables are usually thinner than the cable that should be used.
Now I am using a Toshiba lap top for running the CNC software and driving the mill and I found on testing the motors that I could not get them all running properly, by which I mean not at all in some cases and missing steps in other cases. A lot of time was spent trying different configurations of pin assignments, motor tuning and step pulse width and "active high /active low" settings and the best I could achieve was three of the four axis running reasonably well. In discussion with CNC4PC they offered me their bi-directional breakout board to try as this had more power available to drive the motor controllers. This duly arrived and when tested required the pulse width to be set up with an increase of +5 microsecs and the pins to be "active high" (all set in Mach3) and all the drives worked fine. I understand that the parallel port board is undergoing a redesign to increase its available power output.
............to be continued